Systems and methods for depth-assisted perspective distortion correction

Abstract

Systems and methods for automatically correcting apparent distortions in close range photographs that are captured using an imaging system capable of capturing images and depth maps are disclosed. In many embodiments, faces are automatically detected and segmented from images using a depth-assisted alpha matting. The detected faces can then be re-rendered from a more distant viewpoint and composited with the background to create a new image in which apparent perspective distortion is reduced.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS[0001]The current application claims priority under 35 U.S.C. 119(e) to U.S. Provisional Patent Application Ser. No. 61 / 883,927, entitled “Automatic Depth Assisted Face Perspective Correction for Mobile Device Cameras”, filed Sep. 27, 2013 and U.S. Provisional Patent Application Ser. No. 61 / 949,999, entitled “Depth Regularization and Semiautomatic Matting Using RGB-D Images”, filed Mar. 7, 2014. The disclosures of U.S. Provisional Patent Application Ser. No. 61 / 883,927 and U.S. Provisional Patent Application Ser. No. 61 / 949,999 of which is incorporated herein by reference in their entirety.FIELD OF THE INVENTION[0002]The present invention relates generally to correcting of perspective distortion and more specifically to automatic depth-assisted face perspective distortion correction.BACKGROUND OF THE INVENTION[0003]Close range portraiture photographs, such as self-portraits, are often perceived as having apparent perspective distortions at typi...

AI Technical Summary

Benefits of technology

[0004]Systems and methods in accordance with embodiments of the invention automatically correct apparent distortions in close range photographs that are captured using an imaging system capable of capturing images and depth maps. In many embodiments, faces are automatically detected and segmented from images using a depth-assisted alpha matting. The detected faces can then be re-rendered from a more distant viewpoint and composited with the background to create a new image in which apparent perspective distortion is reduced.

[0011]In yet another embodiment, the image processing application stored in the memory further directs the processor to warp the segmented object image data to create warped object image data by rescaling the warped object image data to correspond in size to the segmented object image data.

[0016]In a still yet further embodiment, the image processing application stored in the memory further directs the processor to store the perspective distortion corrected image data and a registered depth map in an image file.

Problems solved by technology

Close range portraiture photographs, such as self-portraits, are often perceived as having apparent perspective distortions at typical image viewing distances, even if the optics produce a geometrically accurate perspective projection of the scene.

A mismatch in the field of view of the camera and that of the viewing display configuration can result in the perceived distortion, which in portraiture photographs, tends to magnify the size of the nose and chin, among other features.

These distortions are especially common with photographs taken with mobile device cameras due to the wide angular field of view typical of such cameras and close range nature of many self-portraits and candid portraits.

Images